Metalworking in Bronze Age China: The Lost-Wax Process By Peng Pe ...

necessarily conformed to a particular shape or dimension.⁷⁵ As some scholars have observed, if a single spacer was applied, it was most frequently set at the bottom center of the vessel where the mold joins meet; when multiple spacers were used, they often appear around the more angular areas of the finished bronze.⁷⁶ Unique to the section-mold casting of early China, the “spacer” cannot be considered identical to the “chaplet,” a spike-like device that had to be removed after casting, usually associated with lost-wax casting in the Mediterranean world.⁷⁷

After several centuries of development, section-mold casting was further elaborated in the Anyang period (c. 1200–1050 BCE).⁷⁸ With foundry debris excavated on an extensive scale, Anyang workshops are the most thoroughly studied foundry sites of early Bronze Age China. The “consensus view” of Anyang casting, formed through generations of painstaking research, is most precisely summarized by Robert Bagley:

with rare exceptions, the decoration of Anyang bronzes was transferred to the mold from a decorated model. In this procedure, the designer began by making a clay model of the vessel he wished to cast, complete with all its decoration. The mold-maker, who may or may not have been the same person, then used the model to make the fitted sections of a clay mold…The sections were formed one by one against the model, removed from it, and reassembled around a core for casting. To form a section the mold-maker applied fine clay to the model. Perhaps he began by painting on a few layers of clay slip to ensure that the finest details would transfer. After completing the section, he removed it, cut mortises into its edges, let it dry, and replaced it on the model. Then he formed a new section next to the first. The new section would acquire tenons fitting the mortises in the section next to it, guaranteeing a secure fit when the completed mold was assembled for casting.…When bronze was poured into the space between core and mold, the result was a bronze replica of the decorated clay model.⁷⁹

Though this mainstream view has recently been attacked by proposals asserting that models were never decorated, only molds,⁸⁰ and that models were made not of the whole vessel but only of a fraction of it,⁸¹